Upload README.md with huggingface_hub

README.md

@@ -11,7 +11,7 @@ task_categories:
 
 # librispeech_asr_test_vad
 
-A dataset for testing voice activity detection.
+A dataset for testing voice activity detection (VAD).
 
 This dataset uses test splits [`test.clean`, `test.other`] extracted
 from the
@@ -23,16 +23,28 @@ There are two additional features.
 
 2. Binary classification of confidence, called `confidence`.  These binary values [0, 1] are computed as follows.  The default confidence is 1.  After a `speech` transition from 1 to 0 the confidence is set to 0 up to a maximum of three 0s in `speech` (approximately 0.1 second).  This can be used to correct for temporary blips in the `speech` feature and unknown decay in the method under test.
 
-The effective chunk size is 512 audio samples for each `speech` feature.
+This test dataset has little background noise thus enables mixing with noise
+samples to assess voice activity detection robustness.
 
-# License Information
+## Example data
 
-This dataset retains the same license as the source dataset.
+A plot for an example showing audio samples and the `speech` feature.
 
-[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)
+<img src="assets/test_other_item_02.png" alt="Example from test.other"/>
+
+The following example demonstrates short zero blips in the `speech` feature for
+valid short pauses in the talker's speech.  However a VAD model under test may
+have slower reaction time.  The `confidence` feature provides an optional means
+for reducing the impact of these short zero blips when computing metrics for a
+method under test.
+
+<img src="assets/test_clean_item_02.png" alt="Example from test.other"/>
 
 # Example usage of dataset
 
+The model under test must support processing a chunk size of 512 audio samples
+at 16000 Hz generating a prediction for each `speech` feature.
+
 ```console
 import datasets
 import numpy as np
@@ -43,24 +55,17 @@ dataset = datasets.load_dataset("guynich/librispeech_asr_test_vad")
 audio = dataset["test.clean"][0]["audio"]["array"]
 speech = dataset["test.clean"][0]["speech"]
 
-# Compute probabilities from method under test (block size 512).
-speech_probs = method_under_test(audio)
+# Compute probabilities from model under test (block size 512).
+speech_probs = model_under_test(audio)
 
 # Add test code here such as AUC metrics.
 # In practice you would run this across the entire test split.
 roc_auc = roc_auc_score(speech, speech_probs)
-
-# Data for plotting
-time_step = 512 / 16000
-audio_x_ticks = np.linspace(0.0, len(audio) / 16000, len(audio))
-speech_x_ticks = np.linspace(0.0, len(speech) * time_step, len(speech))
-
-# Data for inspecting masked audio with plotting or playback.
-speech_mask = np.repeat(speech, 512)
-masked_audio = audio[:len(speech_mask)] * speech_mask
 ```
 
-The confidence values can be used prior to computing metric
+The confidence values can be used to slice the data.  This removes 6.8% of the
+entire dataset `speech` features and removing these low confidence values
+increases precision.
 ```console
 confidence = dataset["test.clean"][0]["confidence"]
 
@@ -68,31 +73,27 @@ speech_array = np.array(speech)
 speech_probs_array = np.array(speech_probs)
 
 roc_auc_confidence = roc_auc_score(
-    speech_array[confidence],
-    speech_probs_array[confidence],
+    speech_array[np.array(confidence) == 1],
+    speech_probs_array[np.array(confidence) == 1],
 )
 ```
 
-Example plots.
+# Silero-VAD model testing
 
-<img src="assets/test_other_item_02.png" alt="Example from test.other"/>
+Example AUC plots computed for Silero-VAD model model with `test.clean` split.
 
-The following example demonstrates short zero blips in the `speech` feature for
-valid short pauses in the talker's speech.  However a VAD method under test may
-have slower reaction time.  The `confidence` feature provides an optional means
-for reducing the impact of these short zero blips when computing metrics for a
-method under test.
+<img src="assets/roc_test_clean.png" alt="Example from test.clean with Silero-VAD"/>
 
-<img src="assets/test_clean_item_02.png" alt="Example from test.other"/>
+Precision values are increased when data is sliced by confidence values.
+
+<img src="assets/roc_test_clean_exclude_low_confidence.png" alt="Example from test.clean with Silero-VAD"/>
 
-# VAD testing
+# License Information
 
-The VAD method shall supply a voice activity prediction for audio chunks of
-512 samples at rate 16000 Hz.
+This dataset retains the same license as the source dataset.
 
-Example AUC plots computed for `test.clean` split and Silero-VAD model.
+[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)
 
-<img src="assets/roc_test_clean.png" alt="Example from test.clean with Silero-VAD"/>
 
 # Citation Information